Design Trust Based Malware Detection System for Secure Cloud Computing

Abstract

Cloud Computing refers to the use of computer resources as a service on demand via the newlineInternet. It is mainly based on data and applications outsourcing, traditionally stored on users newlinecomputers, to remote servers (data centers) owned, administered, and managed by third parties. Data newlinesecurity is a common concern for all technologies. However, it becomes a major challenge when newlineapplied to an uncontrolled environment like Cloud Computing. It is important to distinguish between newlinethe security risks associated with all IT infrastructures and those introduced by the use of Cloud newlineComputing. These risks are generally associated with open, shared, and distributed environments. newlineTherefore, when analyzing the risks, it is important to separate existing problems from those raised newlineby Cloud Computing. newlineComputing remotely also brings with it the most sensitive and significant aspect of data newlinesecurity. Data being resident on servers and infrastructure that are opaque to users brings with it the newlinehuge advantage of ridding users of all the infrastructure operational liabilities. But this also means newlinethe data owned by users navigates through routes either unknown or no control territory. This newlineglaringly brings the security of data on the cloud with high vulnerability. The first research work newlineexplores dimensionally various such vulnerabilities and related strategies to circumvent them does newlineinvolve a sensitive section of modern technology in today s world of computing. An insight through newlinevarious levels of infrastructure, application, data, and network architecture gives this research newlineenough evidence that the complexities involved in maintaining the security problems of cloud newlinecomputing is nevertheless a path most discussed and debated. newlineDue to diverse software vulnerabilities and hardware attacks, user credentials are vulnerable newlineor could land in demilitarized zones. In the second research work, an attempt is made to explore and newlinefinally propose a trust-based malware detection based on optimal data classification. An improved newlineglowworm swarm opti